Our brain is subject to many interesting biases and fallacies (see ‘You are not so smart‘ for a comprehensive, yet fun, summary), but one has particular implications for the workplace. It’s called the Dunning-Kruger effect: the tendency for some people to substantially overestimate their abilities. In fact what Dunning and Kruger found was that the less competent people are, the more they overestimate their abilities. This is not some sort of ‘fake confidence’, it is better described as ‘honest overconfidence’ – I am confident in my abilities because I really don’t know any better. The result is neatly illustrated in a study of software engineers in Silicon Valley where 40% of respondents believed they were in the top 5% of performers.

So if we really do struggle with an accurate self-assessment of our abilities, what does that mean for management and measuring performance? Well, for a start managers are no better at assessing their managerial and leadership abilities. Dunning found that there is almost no correlation between the self-assessment of managers on their leadership and managerial skills and how they are rated by their peers and their bosses. When they then compared those assessments to objective performance measures, they found that the peers and bosses got it right, it was the self-assessments that bore little relation to reality.

Why do we find it so hard to see our own shortcomings and why do we find it so easy to believe that we are better than we really are?  Dunning suspects that Donald Rumsfeld was onto something when he talked about the ‘unknown unknowns’. He believes that we get our self-assessment wrong because we simply haven’t got access to objective measures of reality. If we can’t even conceive of better ways of doing a task or solving a problem, then we will naturally assume that our own solution is the best there is.

The implication for developing people in the workplace and especially for developing managers and leaders is that left to their own devices most will assume the wrong starting point. If your self-assessment is off the mark you will make the wrong assumptions about what you still need to learn. There is therefore much to be said for giving and receiving performance feedback often. This feedback needs to be on specific, observable behaviour and performance, not about the person or their character. We only become receptive to feedback if it is provided frequently, given in the moment and isn’t coupled to salary or rankings.

The second implication is that true hyperachievers will not be recognised for how good they really are. Given that these top 1% performers can be 10 times more productive than the average worker, companies should be doing everything in their power to recognise them and keep them. Because the average manager will not be able to see this difference in performance objectively, it is likely that without introducing more objective ways of identifying hyperachievers, they will go unrecognised and undervalued. In his studies Dunning found that on average managers had no problem in reliably identifying their poor performers, but he could get no agreement on who the top performers were. We have encountered this scenario frequently and have coached many hyperachievers who often were seen more as troublemakers than assets for their organisations.

These insights into performance and overconfidence are quite new, the studies quoted started in 2003 and some of the results I mentioned were only published in 2011. It would appear we still have a lot to learn about the true link between confidence and competence. Recent work has focussed on the differences between men and women and they have found that men tend to display higher levels of ‘honest overconfidence’. They also found that we ‘buy’ honest overconfidence – because the person isn’t faking the overconfidence (they just don’t know any better), it comes across as genuine and we simply believe the message and the messenger. The result is that low competence coupled with a high level of ‘honest overconfidence’ can get a person promoted, a scenario any one of you will be familiar with. But at least now you understand how it comes about!